DESCRIPTION (Applicant's Abstract): The APOE-4 allele, found in approximately 25% of older adults, is a susceptibility polymorphism for a number of adverse health outcomes, especially Alzheimer's disease (AD). While recognized as a means of increasing the accuracy of diagnosis of AD, the identification of this susceptibility polymorphism provides an opportunity to address important public health questions most relevant to advances in molecular population genetics, pharmacogenetics and public health-- examination of the long-term impact of being positive for E-4 upon quality of life (QOL), and health services use. Using data from a ten-year longitudinal study of adults age 65-105 (EPESE-Duke) we will examine 2,031 community representative persons, equally distributed between African-Americans and Whites, genotyped at the six-year follow-up (32% positive for E-4). QOL is operationally defined to include the following domains: 1) physical health, 2) functional status, 3) cognitive status, 4) psychological well-being, 5) social support, 6) religious/spiritual status, and 7) mortality/active life expectancy. Health service use was determined by self-report and Part A Medicare data. The underlying hypothesis is that only some of the trajectories of QOL and health service use vary between persons with the E-4 allele and those without. Multivariate models will be used to estimate the effects of being E-4 positive on QOL indicators and health services use controlling for demographic characteristics and other potential confounders. Cox proportional hazard, discrete time survival, and hierarchical linear modeling will be used according to the metric and distribution of the dependent variable.